C28 JOURNAL OF FOOD SCIENCE—Vol. 71, Nr. 1, 2006Published on Web 1/10/2006

© 2006 Institute of Food TechnologistsFurther reproduction without permission is prohibited

C: Food Chemistry & Toxicology

JFS C: Food Chemistry and Toxicology

Lactic Acid Fermentation ReducesAcrylamide Formation and OtherMaillard Reactions in French FriesPERNILLE BAARDSETH, HANS BLOM, GRETE SKREDE, LIV T. MYDLAND, ANDERS SKREDE, AND ERIK SLINDE

ABSTRACT: Blanched and nonblanched potato rods (var. Beate) were fermented with Lactobacillus plantarumstrain NC8 (109 colony-forming units [CFU]/mL) at 37 °C for 45 and 120 min. Potato rods were pre-fried at 170 °Cfor 3 min, cooled, and subsequently deep-fried for 2 min 15 s. Potato juice (var. Beate) was fermented with the samestrain (108 CFU/mL) at 30 °C for 1 to 5 h. Lactic acid fermentation of nonblanched potato rods for 45 min reducedacrylamide level in French fries with 48%, and with 71% after 120 min. By blanching potato rods before fermenta-tion, reductions in acrylamide after 45 min and 120 min were 79% and 94%, respectively. Blanching, and especiallyfermentation, reduced visually judged browning of the French fries. Fermentation of potato juice reduced pH from5.70 to 4.05 after 3 h. Simultaneously, glucose declined from 610.8 mg/100 mL to 7.9 mg/100 mL, fructose from457.8 mg/100 mL to 0.0 mg/100 mL, and sucrose from 132.0 mg/100 mL to 29.2 mg/100 mL. Asparagine contentremained largely unaffected between 0 h (1217.5 mmol/100 mL) and 4 h (1175.6 mmol/100 mL) and increasedslightly (1470.3 mmol/100 mL) after 5 h fermentation. Levels of several other amino acids involved in Maillardreactions, that is, alanine, arginine, phenylalanine, and serine, decreased during fermentation. It is concluded thatacrylamide formation during production of French fries can be effectively lowered by lactic acid fermentation ofpotato rods before deep-frying. The reduction is due to reduced levels of reducing sugars rather than reduction ofavailable asparagine.

Keywords: acrylamide, Lactobacillus plantarum NC8, French fries, reducing sugars, amino acids

Introduction

Acrylamide and other Maillard products are formed at high foodprocessing temperatures (Mottram and others 2002; Stadler

and others 2002). Many of the reaction products formed are visibleto the naked eye as brown coloration of products that are fried oroven-baked, such as crisp bread, processed potato products, cook-ies, and coffee (Svensson and others 2003). One of the major routesfor acrylamide formation is the reaction between asparagine andthe carbonyl group of a reducing sugar such as fructose or glucose(Mottram and others 2002; Yaylayan and others 2003; Zyzak andothers 2003; Taeymans and others 2004).

Acrylamide is known to be a neurotoxic, genome-affecting, andcarcinogenic compound (Friedman 2003; Dybing and others 2005).The long-term effects of acrylamide are not yet fully understood.Still, the food industry has recently devoted great attention to un-derstand the chemistry of the compound, to improve the analyti-cal methods, as well as to modify the processing conditions to re-duce formation of acrylamide in food commodities (Taeymans andothers 2004). Because of the discovery of acrylamide in food in 2002,the Food and Drug Administration (FDA) has initiated a broadrange of activities on acrylamide, including being at the forefrontof new toxicology research on acrylamide. The results from thesestudies are expected in 2007. Until then, the FAO/WHO expert com-mittee on Food Additives in February 2005 (JECFA/64/SC) recom-

mended that the efforts toward reduced acrylamide concentrationsin food should continue.

Food processing approaches to reduce the amount of acryla-mide formed could be directed toward reducing the amount of re-actants, that is, asparagine and reducing sugars, and/or makingconditions for the reaction unfavorable. Thus, blanching of pota-toes before frying, lowering pH, and increasing moisture have beenshown to reduce acrylamide levels (Jung and others 2003; Kita andothers 2004; Pedreschi and others 2005).

Lactic acid bacteria convert reducing sugars in vegetables to lac-tic acid thus lowering pH (Slinde and others 1993; Kaaber and oth-ers 1995). Previous studies have shown that lactic acid fermenta-tion of potato (Kaaber and others 1995) and carrot slices (Slindeand others 1993; Baardseth and others 1995, 1996) reduces sugarlevels, amounts of Maillard products, and the burnt taste of deep-fried chips. Preliminary unpublished experiments with deep-friedpotato and carrot chips in our lab have shown that fermentation ofthe slices with Lactobacillus plantarum NC8 may lower acrylamideformation in the final product. However, to the authors’ knowl-edge, no previous study has been published on the effect of lacticacid fermentation on acrylamide contents in heat-processed potatoproducts.

The aim of the present study was to investigate the potential oflactic acid fermentation of potato rods before deep-frying as an in-dustrial process for reducing acrylamide formation and brown colordevelopment in French fries during deep-frying. A further purposewas to investigate whether a reduction in acrylamide was due toreduced levels of sugars alone or if the selected L. plantarum strainalso metabolizes asparagine and other amino acids during the fer-mentation process.

MS 20050332 Submitted 6/2/05, Revised 7/6/05, Accepted 10/14/05. AuthorsBaardseth, Blom, and G. Skrede are with Matforsk AS Norwegian Food Re-search Inst., Osloveien 1, N-1430 Ås, Norway. Authors Mydland and A. Skredeare with Aquaculture Protein Centre, Centre of Excellence, Ås, Norway.Author Slinde is with Inst. of Marine Research, Bergen, Norway. Direct in-quiries to author Baardseth (E-mail: pernille.baardseth@matforsk.no).

Vol. 71, Nr. 1, 2006—JOURNAL OF FOOD SCIENCE C29URLs and E-mail addresses are active links at www.ift.org

C: Fo

od Ch

emist

ry &

Toxic

ology

Reduction of acrylamide by fermentation . . .

Materials and Methods

Bacterial inoculumLactobacillus plantarum strain NC8, a stable plasmid free strain

(Shrago and others 1986; Aukrust and Blom 1992), was grown at37 °C overnight in (MRS) (DIFCO, Detroit, Michigan, U.S.A.), medi-um and harvested at a maximum density of 109 colony-formingunits (CFU)/mL. The cells were centrifuged at 4000 × g for 10 min ina HERAEUS 4KR, using the 1000-mL buckets (Thermo ElectronCorp., Boston, Mass., U.S.A.). The pellet was dissolved in water at37 °C to 109 CFU/mL before addition to potato rods or juice.

Blanching and fermentation of potato rodsPotatoes (var. Beate) were obtained from a local grocery store.

The potatoes were peeled (abrasive peeling C12P, MachinefabriekEillert B.V., Ulft, The Nederland) and cut into 9 × 9 mm rods (Coupe-Frites, Bron-Coucke S.A., Orcier, France). The rods were kept inwater until further treatment. One batch of potato rods was useddirectly for fermentation, while another was blanched in water at 80°C for 5 min and cooled in water before fermentation. Fermentationwas performed with 1-kg potato rods in 2-L jars together with 1 L cellinoculum at 109 CFU/mL. The rods were incubated at 37 °C andsamples were withdrawn after 45 and 120 min of fermentation,dried by shaking, and immediately deep-fried. Control sampleswere deep-fried without delay after peeling and cutting.

Two parallel portions of 200-g each were deep-fried in palm oil(Denofa AS, Fredrikstad, Norway) at 170 °C in a Nuova Elframo,Model EB (Bergamo, Italy) fryer. The frying regime was as followed:deep-fried (pre-frying) for 3 min, cooling in air for 5 min followed bya 2nd deep-frying for 2 min 15 s. Samples were photographed andduplicate samples analyzed for acrylamide.

Extraction and fermentation of potato juicePeeled potatoes (var. Beate) were processed into juice in a Braun

Multipress type 4290 (Braun GmbH, Kronberg, Germany). The juicewas decanted and stored at –25 °C until the start of the experi-ments. Following thawing at room temperature, the juice (90 mL)was inoculated with 10 mL L. plantarum NC8 (109 CFU/mL), givingan initial bacterial density of 108 CFU/mL. The mixture was incu-bated at 30 °C and samples withdrawn immediately and then ev-ery hour for 5 h. Samples were frozen immediately and stored fro-zen until they were analyzed for pH, sugars, and free amino acids.The experiment was performed in duplicate with separate tubes foreach sampling.

AcrylamideAcrylamide was analyzed using solid-phase extraction before

liquid chromatography/mass spectrometry as described by Youngand others (2004).

The homogenization step was slightly modified as 5 g homoge-nized sample was diluted to 100 mL with water, vigorously shakenfor 30 min, and adjusted to pH 7 to 9 by 0.5 M NaOH. The mixturewas filtered through glass-fiber by suction, before aliquots wereapplied on the preconditioned SPE columns. The final analysis isdone by reversed-phase LC and detected by MS, identified by 2 ionfragments.

pHDuplicate samples of potato juice were carefully thawed on ice

and pH was measured immediately using a WTW pH 340 meter(WTW Wissenschaftlich-Technische Werkstätten GmbH, Weilheim,Germany) equipped with a Hamilton Polilyte Lab electrode(HAMILTON Bonaduz AG, Bonaduz, Switzerland) and a WTW TFK

325 temperature sensor. Calibration was performed with HamiltonDURACALTM pH buffers.

SugarsSucrose, glucose, and fructose were determined by high-perfor-

mance liquid chromatography (HPLC) using a Hewlett Packard1100 Series chromatography system (Agilent Technologies, Inc.,Palo Alto, Calif., U.S.A.) and a Gilson 132 refractive index detector(Gilson Medical Electronics Inc., Middleton, Wis., U.S.A.). The chro-matographic system was equipped with a Chrompack Carbohy-drates Pb (5 mm) 300 × 7.8-mm inner dia analytical column and apre-column containing the same material (Varian, Inc., Palo Alto,Calif., U.S.A.). The mobile phase was water, and the system wasoperated at 80 °C at a flow of 0.4 mL/min. Sucrose, glucose, andfructose (Chem Service, West Chester, Pa., U.S.A.) were used asexternal standards.

Potato juice samples were diluted (1:1 w/w) with ethanol, kept at65 °C for 60 min, and passed through C-18 SepPack cartridges,which had previously been activated by 2 mL methanol and 5 mLdeionized water (Bach Knudsen 1997). One mL of the eluents wasdried on a heating block until ethanol was evaporated. Sampleswere diluted to proper volume by water and injected onto the chro-matograph.

Free amino acidsThe free amino acids in the potato juices were analyzed by ion

exchange chromatography on a Lithium High Performance Column(Biochrom Ltd., Cambridge, U.K.) in an automated amino acid an-alyzer (Biochrom 20, Biochrom Ltd.), using lithium-based eluentsand post-column derivatization with ninhydrin (Physiological FluidChemical Kit, Biochrom Ltd). Data were analyzed against externalstandards (Sigma amino acid standard solutions: acidics, neutrals,and basics, supplemented with glutamine, all purchased from Sig-ma Chemical, St. Louis, Mo., U.S.A.) using the Chromeleon® Chro-matography Management Software (Dionex Ltd., Surrey, U.K.).Duplicate samples (2 × 1 mL) of the potato juices were centrifugedat 16000 × g for 15 min at 4 °C (Biofuge Fresco, Heraeus Instru-ments, Kendro Laboratory Products GmbH, Hanau, Germany). Ofthe supernatants, 50 mL were diluted with 450 mL 0.2M lithium cit-rate loading buffer, pH 2.2 (Biochrom Ltd.) and microfiltrated (0.2mm Spartan membrane filter, Schleicher & Schuell, Dassel, Germa-ny) before injection (20 mL).

Calculations and statistical analysisSignificant (P # 0.05) differences among means of each sugar

and amino acid were ranked using the Ryan-Einot-Gabriel-Welschmultiple F test (REGWQ) through the MEANS statement in thePROC GLM procedures of SAS (SAS Inst. 1990).

Results and Discussion

Blanching and fermentation of potato rodsBoth blanching and lactic acid fermentation of potato rods before

deep-frying significantly reduced acrylamide formation during theproduction of the French fries (Figure 1).

The fermentation of nonblanched potato rods reduced acryla-mide contents in the French fries with 48% and 71% after 45 minand 120 min of fermentation, respectively. The combined effects ofblanching and fermentation amounted to a decline in acrylamideformation of 79% after 45 min of fermentation and 94% when fer-mentation continued for 120 min.

The acrylamide level of 2400 mg/kg found in the present study inthe French fries produced from potato rods not subjected to

C30 JOURNAL OF FOOD SCIENCE—Vol. 71, Nr. 1, 2006 URLs and E-mail addresses are active links at www.ift.org

C: Food Chemistry & Toxicology

Reduction of acrylamide by fermentation . . .

blanching or fermentation (Table 1), is well within ranges reportedpreviously. Thus, Lingnert and others (2002) indicated that 300 to700 mg/kg is the typical acrylamide range for French fries with 3500mg/kg as an upper extreme value, whereas Jung and others (2003)reported about 1500 mg/kg and Friedman (2003) found values be-tween 200 and 12000 mg/kg.

Blanching of potato rods has previously been shown to reduceacrylamide formation in French fries. Grob and others (2003) re-ported about 50% reduction in acrylamide level of French fries afterblanching of the potato rods before deep-frying. Similar reductions

were reported after blanching potato slices before deep-frying intopotato chips (Pedreschi and others 2005).

The fermentation process had a pronounced effect on the colorof the French fries as the color turned lighter and less brown whenthe potato rods were fermented before frying (Figure 2). The light-ness increased with the extension of the fermentation period.Blanching was also efficient in reducing brown color of the Frenchfries. Even further improvement in color of the French fries wasachieved when the blanched potato rods were fermented beforedeep-frying.

Sensory analysis of the French fries was not carried out in thepresent study. However, although pH was reduced and sugar lev-els declined during fermentation, there was no distinguishabletaste difference between the fermented French fries and the unfer-mented control. In a limited consumer test no comments were giv-en with regard to difference in taste, but a preference for the lightcolored fries was expressed. In a previous study it was shown thatfermentation of carrots induced a fresh, slightly sour taste to carrotchips (Slinde and others 1993). In French fries, lowering of pH bydipping in 1% citric acid for 1 h before frying caused no detectabletaste difference compared with the control, whereas dipping in 2%citric acid promoted a slightly sour taste and harder texture (Jungand others 2003).

Pedreschi and others (2005) reported a lighter color in blanched,deep-fried potato slices than in nonblanched slices. Lactic acidfermentation of potato slices has previously been shown to increaselightness in deep-fried potato chips (Kaaber and others 1995). Inthe latter study, the change in color caused by the fermentationprocess was shown by instrumental color analysis (increased L*values in the L*a*b* color system) as well as by visual judgments.The color of deep-fried potato products arises from the non-enzy-

Figure 1—Reduction of acrylamide formation in Frenchfries as a function of time of fermenting blanched andnonblanched potato rods with Lactobacillus plantarum NC8(109 colony-forming units [CFU]/mL) before deep-frying.Bars show standard deviation of duplicate experiments

Table 1—Effects of lactic acid fermentation on pH and concentrations of sugars and free amino acids in potato juice

Fermentation period at 30 °C

0 ha 1 h 2 h 3 h 4 h 5 h S.E.Mb

pH 5.70ac 4.78b 4.19c 4.05d 3.93e 4.06d 0.0

Sugars (mg/100 mL)Sucrose 132.0ab 298.9a 174.7ab 29.2b 1.4b 2.4b 67.45Glucose 610.8a 10.6b 12.4b 7.9b 0.0b 0.0b 5.92Fructose 457.8a 77.1b 17.3b 0.0b 0.0b 0.0b 25.77

Free amino acids (mmol/100 mL)Alanine ala 225.9a 222.7a 169.9b 147.9c 142.7c 101.4d 1.72Arginine arg 284.1a 223.2b 224.0b 224.9b 198.2c 148.4d 1.17Asparagine asn 1217.5bc 1233.7b 1214.6bcd 1161.4d 1175.6cd 1470.3a 9.67Aspartic acid asp 256.7a 261.2a 225.7b 211.9c 204.4c 172.8d 1.82g-aminobutyric acid gaba 617.3b 676.4a 662.6a 657.1a 647.2ab 412.5c 6.71Glutamic acid glu 55.7f 90.2e 125.4d 141.1c 147.5b 174.4a 0.88Glutamine gln 1253.9c 1351.3b 1275.9c 1192.4d 1187.0d 1455.1a 9.42Glycine gly 84.1a 86.0a 76.0b 64.7c 67.4c 58.5d 1.00Histidine his 40.1a 41.2a 39.4a 35.9b 35.4b 27.8c 0.36Isoleucine ile 129.0a 133.4a 121.2b 108.9c 104.2c 48.4d 0.98Leucine leu 66.7a 61.3b 52.4c 40.1e 43.4d 24.9f 0.24Lysine lys 114.8ab 121.3a 114.4ab 111.8b 110.5b 85.5c 1.33Methionine met 92.9a 96.3a 86.7b 83.6bc 79.7c 55.4d 0.95Phenylalanine phe 86.3b 88.7a 74.5c 64.6d 57.9e 23.9f 0.34Proline pro 58.6b 62.5a 55.4b 45.7c 40.6d 36.3e 0.66Serine ser 123.5a 102.4b 75.9c 43.8d 40.1e 21.0f 0.61Threonine thr 198.9b 207.9a 195.1b 173.3c 179.1c 136.8d 1.65Tyrosine tyr 19.3b 22.0a 17.2c 11.4d 9.9e 2.4f 0.10Valine val 324.0a 332.5a 308.0b 287.6c 275.8c 185.6d 3.15Total free AAd FAA 5249b 5414a 5114c 4808d 4747de 4641e 23.6aIncubation time (h) after the addition of Lactobacillus plantarum NC8 (108 colony-forming units [CFU]/mL potato juice).bPooled standard error of mean (S.E.M).cMeans within each treatment (row) with different letters are different (P # 0.05).dSum of all analyzed free amino acids (cysteine was not detected).

Vol. 71, Nr. 1, 2006—JOURNAL OF FOOD SCIENCE C31URLs and E-mail addresses are active links at www.ift.org

C: Fo

od Ch

emist

ry &

Toxic

ology

Reduction of acrylamide by fermentation . . .

matic Maillard browning reaction, involving sugars and amino acidson the surface of the potato, with reducing sugars being the limit-ing factor (Marquez and Añón 1986). Pedreschi and others (2005)reported a linear correlation between acrylamide content of thepotato chips and their color represented by the redness compo-nent a* in a given temperature range (120 °C to 180 °C). With themechanisms revealed for acrylamide formation, that is, reactionsbetween a reducing sugar and asparagine (Yaylayan and others2003; Zyzak and others 2003), the industry’s efforts to producelight-colored potato products coincide with their requirements forproducing products low in acrylamide (Becalski and others 2004).However, browning level is not a reliable measure of acrylamidecontent in large-surface fried potato products (Taubert and others2004).

Several strategies have been suggested for keeping reducingsugars in the raw material at low levels, that is, selection of suitablepotato varieties (Amrein and others 2003; Matthäus and others2004; Olsson and others 2004; Williams 2005), optimizing storageregimens to prevent cold-induced production of reducing sugar(Noti and others 2003; Matthäus and others 2004) and blanchingto wash out sugars before deep-frying (Grob and others 2003; Kitaand others 2004; Pedreschi and others 2004). Furthermore, reduc-ing the pH by soaking in acids resulted in a lower rate of Maillardbrowning (Slinde and others 1993; Aukrust and others 1995) andacrylamide formation (Jung and others 2003; Kita and others 2004;Weisshaar 2004). Modifying deep-frying conditions (Becalski andothers 2003; Granda and others 2004) and optical sorting to removedark products (Taeyemans and others 2005) have also been advo-cated.

Extraction and fermentation of potato juiceIt is well known that Lactobacilli metabolize glucose and fructose,

and we have previously shown that these sugars are also metabo-lized during lactic acid fermentation of potato (Kaaber and others1995) and carrot (Slinde and others 1993) slices. The bacterial re-moval of sugars will be limited to the outer layers of the vegetable

material, and the total amount of sugars metabolized will dependon the surface-to-volume ratio of potato slices or rods. We thereforechose potato juice as a model for studying sugar and amino acidlevels during lactic acid fermentation of the raw material.

Levels of reducing sugars in the potato juiceLactic acid fermentation of potato juice reduced pH from 5.70 to

4.19 after 2 h of fermentation and further to 4.05 after fermentingfor 3 h, where it remained stable for the rest of the experimentalperiod (Table 1). The initial sugar levels of the potato juice used inthe present study were comparable, although in the upper range,of levels previously reported for potatoes (Kaaber and others 1995;Becalski and others 2004; Kita and others 2004; Williams 2005). Adecrease in sucrose concentration occurred after 3 h of fermenta-tion. The numerical, but statistically not significant, increase insucrose during the 1st h of fermentation (Table 1) may have beencaused by endogenous invertase activity, which might have hydro-lyzed sucrose during preparation of the nonfermented potato juicebefore the sugar analysis. The contents of glucose and fructosewere greatly reduced already after the 1st h of fermentation, andno monosaccharides were detectable in the juice after 4 to 5 h offermentation.

Kaaber and others (1995) obtained reductions in total sugar con-tents in slices from various potato varieties between 72% and 96%after fermentation for 24 h at 23 °C with Lactobacillus strain NCIMB40450 at an inoculum level of 107 CFU/mL. The present reductionof 97% of total sugars within 3 h of fermentation (Table 1) demon-strates the importance of selecting an efficient bacterial strain, highinoculum density, and optimal temperature for the effectiveness ofbacterial degradation of sugars.

Levels of free amino acids in the potato juiceThere were considerable changes in the levels of free amino acids

(FAA) in the potato juice after lactic acid fermentation (Table 1).Initially, a 3% net increase (P < 0.05) in the total amount of FAA wasobserved after 1 h of fermentation. This might be due to both en-

Figure 2—Browning inblanched and non-blanched, fermentedand unfermented, deep-fried French-fries. Aand D = unfermented (A= blanched, D =nonblanched); B and C= blanched and fer-mented for 45 and 120min; E and F =nonblanched andfermented for 45 and120 min.

C32 JOURNAL OF FOOD SCIENCE—Vol. 71, Nr. 1, 2006 URLs and E-mail addresses are active links at www.ift.org

C: Food Chemistry & Toxicology

Reduction of acrylamide by fermentation . . .

dogenous protease activity in the potato juice and lysis of some ofthe bacterial cells added at the start of incubation. However, sam-ples collected at later stages during fermentation (2 to 5 h) showeda decrease (P < 0.05) in FAA compared with the amounts before fer-mentation. The Lactobacilli fermentation of AA in potato slices orrods will be limited to the outer layers of the vegetable material,thus the total amount of protein/AA metabolized will have a minoreffect on the nutritional quality of the nitrogenous nutrients in thefinal product. The 2 most abundant amino acids present in freeform in all samples were the amides asparagine and glutamine (ap-proximately 25% and 30% of total FAA, respectively). This is in ac-cordance with earlier findings (Brierley and others 1996; Amreinand others 2004; Olsson and others 2004).

In the present study, all amino acids except glutamic acid, aspar-agines, and glutamine showed a marked decrease (P < 0.05) duringthe 5 h of fermentation. The most prominent decrease was ob-served in the concentrations of isoleucine, leucine, phenylalanine,serine, and tyrosine (ranging from 63% to 87% reduction, comparedwith initial values). When growing L. plantarum in culture, valine,isoleucine, leucine, tyrosine, and phenylalanine were consumedmainly as free amino acids, whereas asparagine, proline, lysine,and arginine were mainly derived from peptides (Kask and others1999). In addition, several FAA (asparagine, serine, glutamic acid,valine, isoleucine, leucine, phenylalanine) were consumed in high-er amounts than needed to build up cell protein (Kask and others1999). For growth of L. plantarum NC8 in culture, arginine, leucine,isoleucine, tyrosine, and valine were essential, whereas asparaginewas not (Møretrø and others 1998). Some proteolytic activities areassociated with growth of L. plantarum NC8, as milk is clotted after12 h at 20 °C with a 106 CFU/mL inoculum (Holck and Næs 1992).Upon fermenting potato rods, this may influence proteins on thesurface of the potato rods and provide amino acids for use ingrowth. The practical implementation of this must, however, bejudged from the knowledge that the free amino acids present in thepotato covers ample concentrations of all amino acids essential forgrowth of L. plantarum NC8 (Arg, Leu, Ileu, Tyr, and Val) (Møretrøand others 1998).

In the present experiment, we added a high number of bacteria(108 CFU/mL) to the potato juice, thus only a small increase in cellnumber (up to 109 CFU/mL) is to be expected. However, even thisnear to steady-state condition requires amino acids for growth andmaintenance.

When comparing our results of changes in the FAA concentra-tions in the potato juice during the fermentation with L. plantarumNC8, with the amino acid composition of the L. plantarum biomassreported by Kask and others (1999), it seems that the L. plantarumNC8 strain consumed more arginine, serine, phenylalanine, me-thionine, isoleucine, valine, leucine, tyrosine, glx (glutamic acid[glu] + glutamine [gln]) and asx (aspartic acid [asp] + asparagine[asn]), and less lysine, proline, glycine, histidine, and threoninethan needed to build up cell protein. Between 4 and 5 h of fermen-tation, the concentrations of asparagine and glutamine increased(P < 0.05) by approximately 24%, whereas the concentration of g-aminobutyric acid was reduced (P < 0.05) by 36%. Simultaneously,the pH of the potato juice (Table 1) also showed an increase (P <0.05). This might be a result of the bacterial cells trying to compen-sate and adjust to the nonphysiological low pH values (<4) and alsopossibly due to increased hydrolysis of peptides/proteins in thepotato juice. Lactobacilli have been shown to possess a specificantiport for glutamate/g-aminobutyric acid (Higuchi and others1997; Cotter and Hill 2003). It is not known, however, whether theantiport can reverse this transport across the cellular membrane. Itis worth bearing in mind, however, that as the pH in the potato juice

decreases (that is, due to the fermentation and release of lacticacid), the free amino groups become protonated and some FAA(that is, serine, valine, and isoleucine) are less stable at lower pHvalues.

It is well established that acrylamide is formed from asparagineand reducing sugars like glucose and fructose by the Maillard reac-tion occurring at high temperatures (Mottram and others 2002; Sta-dler and others 2002). Fructose may be more reactive in acrylamideformation than glucose (Becalski and others 2004). The rearrange-ment and degradation of a Schiff base to acrylamide is at presentnot fully understood and neither is the kinetics of the reaction be-tween carbonyl and asparagine (Taeymans and others 2004). Fur-thermore, the key intermediates of the reaction remains to be elu-cidated (Dybing and others 2005). In the 1st experiment of thepresent study, we showed that including a lactic acid fermentationstep before the deep-frying of French fries could reduce acryla-mide formation. In the 2nd experiment, we revealed that the con-centration of both glucose and fructose declined rapidly duringlactic acid fermentation, whereas asparagine contents remainedlargely unaffected during the 1st h after incubation. This indicatesthat reducing sugar is a key rate-limiting factor for acrylamide for-mation in fried potato products such as French fries. This is in agree-ment with the recent studies by Amrein and others (2004), reveal-ing that acrylamide formation in potato tubers was mainlydetermined by the contents of glucose and fructose and not by thecontent of asparagine. Conversely, Zyzak and others (2003) andBecalski and others (2004) found that the level of asparagine af-fected content of acrylamide in certain potato products. Lactobacil-lus plantarum NC8 shows no asparaginase activity, although otherLactobacillus species are capable of hydrolyzing asparagine to as-partic acid (Møretrø and others 1998).

The decreased levels of reducing sugars most likely caused theobserved reduction in thermal browning during deep-frying ofpotato rods. The fermentation also caused substantial reduction inlevels of some of the amino acids shown to be involved in modelMaillard reactions, especially alanine, arginine, phenylalanine, andserine (Ajandouz and Puigserver 1999; Kwak and Lim 2004). Therewere only small changes in lysine content during the 1st 3 to 4 h offermentation, and lysine is known to be the most reactive aminoacid in Maillard reactions (Adrian 1974).

Conclusions

Acrylamide formation during production of French fries can beefficiently lowered by lactic acid fermentation of the potato

rods before deep-frying. The reduction in the formation of acryla-mide by fermenting the potato rods with L. plantarum NC8 beforedeep-frying is due to reduction in the levels of reducing sugar ratherthan a specific reduction in the levels of available asparagine. Lac-tic acid fermentation also reduces formation of Maillard productsclearly visible as reduced brown color of the French fries.

AcknowledgmentsWe would like to acknowledge Forinnova AS, Bergen, Norway forfinancial support.

ReferencesAdrian J. 1974. Nutritional and physiological consequences of the Maillard re-

action. World Rev Nutr Diet 19:71–122.Ajandouz E, Puigserver A. 1999. Nonenzymatic browning reaction of essential

amino acids: effect of pH on caramelization and Maillard reaction kinetics. JAgric Food Chem 47:1786–93.

Amrein TM, Bachmann S, Noti A, Biedermann M, Barbosa MF, Biedermann-Brem S, Grob K, Keiser A, Realini P, Escher F, Amadò R. 2003. Potential of acry-lamide formation, sugars, and free asparagine in potatoes: a comparison ofcultivars and farming systems. J Agric Food Chem 51:5556–60.

Amrein TM, Schönbächler B, Rohner F, Lukac H, Schneider H, Keiser A, Escher F,

Vol. 71, Nr. 1, 2006—JOURNAL OF FOOD SCIENCE C33URLs and E-mail addresses are active links at www.ift.org

C: Fo

od Ch

emist

ry &

Toxic

ology

Reduction of acrylamide by fermentation . . .

Amadò R. 2004. Potential for acrylamide formation in potatoes: data from the2003 harvest. Eur Food Res Technol 219:572–8.

Aukrust T, Blom H. 1992. Transformation of Lactobacillus strains used in meatand vegetable fermentations. Food Res Int 25: 253–61.

Aukrust TW, Blom H, Slinde E. 1995. Influence of brine composition on yield andquality of deep-fried fermented carrot chips. Lebensm Wiss Technol 28:100–4.

Baardseth P, Rosenfeld HJ, Sundt TW, Skrede G, Lea P, Slinde E. 1995. Evaluationof carrot varieties for production of deep-fried carrot chips. I. Chemical as-pects. Food Res Int 28:195–200.

Baardseth P, Rosenfeld HJ, Sundt TW, Skrede G, Lea P, Slinde E. 1996. Evaluationof carrot varieties for production of deep fried carrot chips. II. Sensory aspects.Food Res Int 28:513–9.

Bach Knudsen KE. 1997. Carbohydrate and lignin content of plant materialsused in animal feeding. Anim Feed Sci Technol 67:319–38.

Becalski A, Lau BPY, Lewis D, Seaman SW. 2003. Acrylamide in foods: occurrence,sources, and modeling. J Agric Food Chem 51:802–8.

Becalski A, Lau BPY, Lewis D, Seaman SW, Hayward S, Sahagian M, Ramesh M,Leclerc Y. 2004. Acrylamide in French fries: influence of free amino acids andsugars. J Agric Food Chem 52:3801–6.

Brierley ER, Bonner PLR, Cobb AH. 1996. Factors influencing the free aminoacid content of potato (Solanum tuberosum L.) tubers during prolonged stor-age. J Sci Food Agric 70:515–25.

Cotter P, Hill C. 2003. Surviving the acid test: responses of gram-positive bacteriato low pH. Microbiol Molec Biol Rev 67:429–53.

Dybing E, Farmer PB, Andersen M, Fennell TR, Lalljie SPD, Müller DJG, Olin S,Petersen BJ, Schlatter J, Scholz G, Scimeca JA, Slimani N, Törnqvist M, Tuijte-laars S, Verger P. 2005. Human exposure and internal dose assessments ofacrylamide in foods. Food Chem Toxicol 43:365–410.

[FAO/WHO] Food and Agriculture Organization/World Health Organization.2005. Expert committee (JECFA/64/SC) on food additives—toxicological in-take evaluations and recommendations on specific contaminants. Rome, Ita-ly: FAO/WHO]

Friedman M. 2003. Chemistry, biochemistry, and safety of acrylamide. A review.J Agric Food Chem 51:4504–26.

Granda C, Moreira RG, Tichy SE. 2004. Reduction of acrylamide formation inpotato chips by low-temperature vacuum frying. J Food Sci 69:E405–11.

Grob K, Biedermann M, Biedermann-Brem S, Noti A, Imhof D, Amrein T, Pfeffer-le A, Bazzocco D. 2003. French fries with less than 100 mkg acrylamide. a col-laboration between cooks and analysts. Eur Food Res Technol 217:185–94.

Higuchi T, Hayashi H, Abe K. 1997. Exchange of glutamate and g-aminobutyratein a Lactobacillus strain. J Bacteriol 179:3362–4.

Holck A, Næs H. 1992. Cloning, sequencing and expression of the gene encodingthe cell-envelope-associated proteinase from Lactobacillus paracasei subsp.paracasei NCDO151. J Gen Microbiol 138:1353–64

Jung MY, Choi DS, Ju JW. 2003. A novel technique for limitation of acrylamideformation in fried and baked corn chips and in French fries. J Food Sci 68:1287–90.

Kaaber L, Sundt TW, Slinde E. 1995. Production of chips from potato cultivars(Solanum tuberosum L.) with high sugar content using lactic acid fermenta-tion. Potato Res 38:39–45.

Kask S, Laht TM, Pall T, Paalme T. 1999. A study on growth characteristics andnutrient consumption of Lactobacillus plantarum in A-stat culture. Antonievan Leeuwenhoek 75:309–20.

Kita A, Bråthen E, Knutsen SH, Wicklund T. 2004. Effective ways of decreaingacrylamide content in potato crisps during processing. J Agric Food Chem52:7011–6.

Kwak E, Lim SI. 2004. The effect of sugar, amino acids, metal ion, and NaCl onmodel Maillard reaction under pH control. Amino Acids 27:85–90.

Lingnert H, Grivas S, Jägerstad M, Skog K, Törnqvist M, Åman P. 2002. Acryla-

mide in food: mechanisms of formation and influencing factors during heat-ing of foods. Scand J Nutr 46:159–72.

Marquez G, Añon MC. 1986. Influence of reducing sugars and amino acids in thecolor development of fried potatoes. J Food Sci 51:157–60.

Matthäus B, Haase NU, Vosmann K. 2004. Factors affecting the concentration ofacrylamide during deep-fat frying of potatoes. Eur J Lipid Technol 106:793–801.

Mottram DS, Wedzicha BL, Dodson AT. 2002. Acrylamide is formed in the Mail-lard reaction. Nature 419:448.

Møretrø T, Hagen BF, Axelsson L. 1998. A new, completely defined medium formeat lactobacilli. J Appl Microbiol 85:715–22.

Noti A, Biedermann-Brem S, Biedermann M, Grob K, Albisser P, Realini P. 2003.Storage of potatoes at low temperature should be avoided to prevent increasedacrylamide formation during frying or roasting. Mitt Lebensm Hyg 94:167–80.

Olsson K, Svensson R, Roslund CA. 2004. Tuber components affecting acryla-mide formation and color in fried potato: variation by variety, year, storagetemperature and storage time. J Sci Food Agric 84:447–58.

Pedreschi F, Kaack K, Granby K. 2004. Reduction of acrylamide formation inpotato slices during frying. Lebensm Wiss Technol 37:679–85.

Pedreschi F, Moyano P, Kaack K, Granby K. 2005. Color changes and acrylamideformation in fried potato slices. Food Res Int 38:1–9.

SAS Inst. 1990. Statistical analysis system user’s guide. Vol. 1, 2. Ver. 6. 4th ed.Cary, N.C.: SAS Inst.

Shrago AW, Chassy BM, Dobrogosz WJ. 1986. Conjugal plasmid transfer (pAMb1)in Lactobacillus plantarum. Appl Environ Microbiol 52:574–6.

Slinde E, Skrede G, Aukrust T, Blom H, Baardseth P. 1993. Lactic acid fermenta-tion influence on sugar content and color of deep-fried carrot chips. Food ResInt 26:255–60.

Stadler RH, Blank I, Varga N, Robert F, Hau J, Guy PA, Robert MC, Riediker S. 2002.Acrylamide from Maillard reaction products. Nature 419:449–50.

Svensson K, Abramsson L, Becker W, Glynn A, Hellenäs KE, Lind Y, Rosèn J. 2003.Dietary intake of acrylamide in Sweden. Food Chem Toxicol 41:1581–6.

Taeymans D, Wood J, Ashby P, Blank I, Studer A, Stadler RH, Gondè P, van Eijck P,Lalljie S, Lingnert H, Lindblom M, Matissek R, Müller D, Tallmadge D, O’BrienJ, Thompson S, Silvani D, Whitmore T. 2004. A review of acrylamide: an indus-try perspective on research, analysis, formation, and control. Crit Rev FoodSci Nutr 44:323–47.

Taeymans D, Andersson A, Ashby P, Blank I, Gondè P, van Eijck P, Faivre V, LalljieSPD, Lingnert H, Lindblom M, Matissek R, Müller D, Stadler RH, Studer A,Silvani D, Tallmadge D, Thompson G, Whitemore T, Wood J, Zyzak D. 2005.Acrylamide: update on selected research activities conducted by the Europeanfood and drink industry. J AOAC Int 88:234–41.

Taubert D, Harlfinger S, Henkes L, Berkels R, Schömig E. 2004. Influence of pro-cessing parameters on acrylamide formation during frying of potatoes. J Ag-ric Food Chem 52:2735–9.

Weisshaar R. 2004. Acrylamide in heated potato products—analytics and forma-tion routes. Eur J Lipid Technol 106:786–92.

Williams JSE. 2005. Influence of variety and processing conditions on acryla-mide levels in fried potato crisps. Food Chem 90:875–81.

Yaylayan V, Wnorowski A, Locas CP. 2003. Why asparagine needs carbohydratesto generate acrylamide. J Agric Food Chem 51: 1753 - 1757.

Young, MS, Jenkins, KM, Mallet, CR 2004. Solid-phase extraction and cleanupprocedures for determination of acrylamide in fried potato products by liquidchromatography/mass spectrometry. J AOAC Int 87:961–4.

Zyzak DV, Sanders RA, Stojanovic M, Tallmadge DH, Eberhart BL, Ewald DK,Gruber DC, Morsch TR, Strothers MA, Rizzi GP, Villagran MD. 2003. Acrylamideformation mechanism in heated foods. J Agric Food Chem 51:4782–7.